Research Report
Office of Institutional Research
Report on Plus and Minus Grading
Fall 2006
This report summarizes the findings of the plus and minus grading trial conducted at
Western Kentucky University, which began in the spring semester of 2005 and continued
through the summer of 2006 (including winter term of 2006). During the trial period, faculty
had the option of recording student grades with a plus or minus sign to better differentiate
student performance. Throughout the grading trial, the data were maintained in the central
university computer system then extracted for analysis at the conclusion of the trial period.
Student GPAs were computed using both signed and unsigned grades. The difference in
these two calculations was contrasted to evaluate the overall effect of signed grading on
students GPAs.
To examine the effect of signed grades (plus/minus grades), two GPA values were
computed for each student. One GPA was computed from quality points derived from signed
grades and the other GPA was derived from quality points derived from unsigned grades.
Western’s administrative computing services assigned quality point values to each grading
method, then computed GPAs as part of each student's data record. Table 1 displays the
quality points assigned with each method.
Grading Method
Letter
Grade
Assigned
A+
A
AB+
B
BC+
C
CD
F
Traditional Grading
Quality Points
4.00
3.00
2.00
1.00
0.00
Signed Grading
Quality Points
4.00
4.00
3.67
3.33
3.00
2.67
2.33
2.00
1.67
1.0
0.0
Table 1. Letter Grade and Associated Quality Points By Grading Method
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
The difference in the two grading methods were measured by comparing the numeric
values of the two GPAs:
GPA Difference = ((Signed Grading GPA) - (Traditional GPA))
A negative value indicates that the signed grading had the net effect of lowering the
GPA. A positive value indicates the signed grading increased GPA. A zero value indicates
no differences in GPA between the two grading methods.
Study Limitations:
During the six semesters of this grading trial, the registrar recorded 222,897 grades.
Two data files were used to store the data:
(1) An original file of unsigned grades (official Banner file used in grade reporting)
(2) A secondary file of signed grades (unofficial grade file used for this study)
A significant number of unsigned grades were changed in the official data file
through the change of grade process. The secondary grade data file did not keep up with the
grade changes in the official file. As a result not all originally signed grades could be
matched with changed official grades. This resulted in some signed grades being excluded
from the study.
A second limitation was the low number of faculty who submitted signed grades
during the trial period. Approximately 33% of all faculty submitted signed grades each term
(Table 6). This 33% translates to about 35% of courses each term being signed graded (Table
7). This low participation rate limits estimating the true effect of signed grading on GPAs
because of the relatively low number of courses graded with signed grading.
A third limitation of this study is related to the number of signed grades a single
student could receive during the grading trial. Because faculty participation was voluntary, a
student may not have received many signed grades and consequently see little difference
between signed and un-signed GPA values. The combination of voluntary faculty
participation coupled with only six terms of grading trial likely minimized the influence of
signed grading on GPA.
A fourth limitation of this study is that the cumulative GPA includes semesters of
traditional grading that occurred prior to the study trial. Measuring the full effect of signed
grading on cumulative GPA is severely clouded by the inclusion of non-signed grades.
Page 2 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Report Sections
Section I: Distribution Of Grades
(Table 2)
(Table 3)
(Table 4)
(Table 5)
Overall Distribution of Grades From Signed Sections
Grade Sign Distribution From Signed Sections By Term
Grade Distribution From Signed Sections By Course Credit Hours
Grade Distribution From Signed Sections By Course College
Section II: Faculty Participation
(Table 6) Faculty Grading Method By Term
(Table 7) Distribution Of Courses By Term By Grading Method
Section III: Effect of Signed Grading On Student GPA
(Table 8) GPA Gain and Loss Percentages
(Table 9) Statistical Changes In Term GPA
(Table 10) Cumulative GPA Gain and Loss Percentages
(Table 11) Statistical Changes In Cumulative GPA
(Table 11b) Change In Sub-Group Term GPA Relative To Signed Courses Enrolled
(Table 11c) Change In Sub-Group Cumulative GPA Relative To Signed Courses Enrolled
(Table 12) Gain/Loss Percentages In Cumulative GPA By College
Section IV: Effect Of Signed Grading on Specific Student Groups
-Students With 4.0 GPAs
-Candidates For Graduation
-Student Receiving Financial Aid
-Student Athletes
-Honors Students
Section V: Grade Distribution Comparison
Section VI: Study Summary
Page 3 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Section I: Distribution Of Grades
Over the six semesters of this trial, the registrar recorded a total of 222,290 grades. Of
that number, 83,004 (37%) were from 4,338 course sections using signed grading. For the
purposes of this study a course was judged to be a ‘signed grading course’ if any grade in the
course was signed. Conversely, a course was judged ‘traditional’ if no grade in the course
was signed. The 4,338 signed sections served as the basis for analysis in this study.
Table 2 provides the distribution of grades from all signed sections, over the six terms
of the trial.
Grade
A+
A
AB+
B
BC+
C
CD
F
Frequency
5,162
17,307
9,802
5,533
12,922
6,759
3,258
7,749
3,653
4,917
5,942
Percent Of All
Grades
6.2
20.8
11.8
6.6
15.5
8.1
3.9
9.3
4.4
5.9
7.1
Cumulative
Frequency
5,162
22,469
32,271
37,804
50,726
57,485
60,743
68,492
72,145
77,062
83,004
Cumulative
Percent
6.2
27.0
38.8
45.4
60.9
69.1
73.0
82.2
86.6
92.5
100.0
Percent Within
Grade Level
16.0
52.6
30.3
21.9
51.2
26.8
22.2
52.8
24.9
100.0
100.0
Table 2. Overall Distribution of Grades From Signed Sections
With the exception of winter term 2006 (which had a higher proportion of plus
grades), the percentages in Table 2 parallel those for each semester of the trial. The
percentage of minus grades at each grade level is consistently higher than that of the plus
grades. The number of ‘A-s’ is about double that of ‘A+s’, the number of ‘B-s’ is almost
25% higher than the ‘B+s’, while the number of ‘C+s’ is only marginally lower than
the ‘C-s.’
Table 3 (next page) summarizes the number and percentage of plus, minus and
unsigned grades reported during the grading trial period. Data from this table suggest that
the outcome of signed grading on GPAs is more negative than positive.
Page 4 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Plus Grades
Term
Minus Grades
%
N
%
6,500
25.11
15,109
58.36
19.09
971
26.04
2,046
54.87
4,398
16.62
6,370
24.08
15,690
59.30
Winter 2006
101
21.22
80
16.81
295
61.97
Spring 2006
3,893
16.79
5,511
23.77
13,783
59.44
569
17.43
782
23.95
1,914
58.62
13,953
16.81
20,214
24.35
48,837
58.84
Spring 2005
N
%
4,280
16.53
712
Summer 2005
Fall 2005
Summer 2006
Total
N
Not Signed
Table 3. Grade Sign Distribution From Signed Sections By Term
Table 4 summarizes the grade distribution by course credit. To conserve space, credit
hours between 0.5 and 1.5 have been grouped, as have courses with credit hours beyond 6.
As would be expected, the distribution of grades in 3-hour courses (the majority of WKU’s
courses) parallels that of the overall distribution (Table 2) with fewer ‘A+’ and ‘B+’ grades
than ‘A-’ or ‘B-’grades. Courses at 6+ credit hours had a majority of grades at the "A+"
level.
Course Credit Hours
0.5-1.5
Grade
2.0
3.0
N
%
N
%
N
522
12.8
351
13.0
1,467
36.0
A-
477
B+
4.0
4.5
5.0
%
N
%
N
%
4,158
5.7
100
4.2
15
3.6
.
859
31.8 14,615
20.0
296
12.4
54
12.9
11
6.4
5
19.2
11.7
391
14.5
8,629
11.8
256
10.8
34
8.1
12
6.9
3
11.5
235
5.8
190
7.0
4,976
6.8
111
4.7
12
2.9
8
4.6
1
3.8
B
495
12.1
250
9.3 11,683
16.0
400
16.8
61
14.6
33
19.1
.
.
B-
211
5.2
161
6.0
6,091
8.3
234
9.8
37
8.9
24
13.9
1
3.8
C+
95
2.3
83
3.1
2,977
4.1
92
3.9
8
1.9
3
1.7
.
.
C
232
5.7
104
3.9
7,082
9.7
247
10.4
57
13.6
27
15.6
.
.
C-
73
1.8
71
2.6
3,275
4.5
185
7.8
29
6.9
20
11.6
.
.
D
101
2.5
84
3.1
4,475
6.1
203
8.5
42
10.0
12
6.9
.
.
F
171
4.2
156
5.8
5,269
7.2
254
10.7
69
16.5
23
13.3
.
.
A+
A
All
N
6+
%
N
. 16
%
61.5
4,079 100.0 2,700 100.0 73,230 100.0 2,378 100.0 418 100.0 173 100.0 26 100.0
Table 4. Grade Distribution From Signed Sections By Course Credit Hours
Page 5 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Table 5 displays the distribution of signed grades by college where the course was
offered. What is apparent from this table is the disproportionate number of “A-” grades
relative to “A+” grades awarded within the Arts and Letters College. Better than three times
the number of “A-” were awarded than “A+.” This was the largest differential within the
university at the ‘A’ grade level.
As a percentage of total college grades, Gordon Ford College of Business and the
Science and Engineering College awarded the smallest proportion of “A+” and “B+” grades.
College Where Course Was Offered
Community
College
Gordon
Ford
College of
Business
Science &
Engineering
Health &
Human
Services
Arts &
Letters
N
%
N
%
N
%
N
573
7.2
249
4.0
686
4.8
1,447
12.7 1,073
17.2
2,229
15.7
%
Education
&
Behavioral University
Science
College
N
%
N
%
N
%
4.1
1,159
11.6
939
11.7 102
11.2
6,845
19.2
2,892
28.9 2,908
36.2 352
38.6
14.0 1,080
13.4
82
9.0
Grade
A+
A
1,005
A-
777
9.8
628
10.1
1,271
9.0
4,559
12.8
1,404
B+
459
5.8
353
5.7
765
5.4
2,743
7.7
705
7.0
451
5.6
55
6.0
B
879
11.1 1,130
18.1
2,443
17.2
5,933
16.6
1,521
15.2
917
11.4
99
10.9
B-
693
8.7
509
8.1
1,178
8.3
3,261
9.1
644
6.4
442
5.5
32
3.5
C+
351
4.4
231
3.7
595
4.2
1,612
4.5
237
2.4
206
2.6
26
2.9
C
625
7.9
775
12.4
1,857
13.1
3,477
9.7
628
6.3
346
4.3
41
4.5
C-
611
7.7
317
5.1
786
5.5
1,511
4.2
201
2.0
199
2.5
27
3.0
D
708
8.9
555
8.9
1,191
8.4
1,919
5.4
286
2.9
228
2.8
30
3.3
F
1,256
15.8
426
6.8
1,182
8.3
2,368
6.6
327
3.3
317
3.9
66
7.2
Totals 7,937 100.0 6,246 100.0 14,183 100.0 35,675 100.0 10,004 100.0 8,033 100.0 912 100.0
Table 5. Grade Distribution From Signed Sections By Course College
Page 6 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Section II: Faculty Participation
During the six terms of the grading trial approximately 1,100 faculty members
reported grades each fall and spring term, about 450 during summer terms, and about 100
during the 2006 winter term.
Faculty had the option of recording grades using either a signed grade or unsigned
grade. For any course faculty could use any combination of grading:
(1) Exclusively plus/minus grading,
(2) Exclusively traditional grading,
(3) A mixture of both plus/minus and traditional grading.
Table 6 shows the distribution of faculty grading methods by term.
Spring
2005
N
%
Exclusively
Traditional
584
Exclusively
Plus/Minus
Both Traditional
& Plus Minus
Fall
Summer
2005
N
2005
Winter
2006
Spring
2006
Summer
2006
%
N
%
N
%
N
%
N
%
52.7 267
59.3
634
53.9
70
64.2
613
55.2 268
63.2
337
30.4 124
27.6
345
29.3
35
32.1
297
26.7 115
27.1
187
16.9
13.1
197
16.8
4
3.7
201
18.1
Instructor
Grading Method
Totals
59
41
9.7
1108 100.0 450 100.0 1176 100.0 109 100.0 1111 100.0 424 100.0
Table 6. Faculty Grading Method By Term
Over the course of the grading trial about 30% of the faculty opted to use exclusively
signed grading, with the majority (56%) opting for traditional grading. The number of faculty
using both methods was reasonably constant at about 16%.
Table 7 reports the distribution of courses, by term, and their associated grading.
Most courses were graded using traditional letter grades rather than signed grades. Initially,
about 37% of all courses offered were signed graded. Toward the end of the trial period the
number of courses being sign-graded had fallen to about 28%.
Page 7 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Course Grading Method
Plus/Minus
Total
Courses
Traditional
Term
N
%
N
%
N
Spring 2005
1,302
36.98
2,219
63.02
3,521
302
29.99
705
70.01
1,007
1,290
35.08
2,387
64.92
3,677
Winter 2006
40
31.50
87
68.50
127
Spring 2006
1,148
32.32
2,404
67.68
3,552
256
28.19
652
71.81
908
4,338
33.91
8,454
66.09
12,792
Summer 2005
Fall 2005
Summer 2006
All
Table 7. Distribution Of Courses By Term By Grading Method
Page 8 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Section III: Effect of Signed Grading On Student GPAs (Sub-group)
To gauge the effect of signed grading on term GPAs, only students who were in
signed graded sections (n=4,338) were selected for analysis (student n = 43,967). The
difference between sign graded and traditional term GPAs was calculated by subtracting the
signed term GPA from the non-signed term GPA.
The modest number of courses using signed grading makes gauging the full effect of
plus/minus grading on student GPA difficult because of the relatively few signed courses in
any given term that would be included in calculating GPA. Obviously, the more sign courses
included the more likely to see changes in GPA.
Table 8 summarizes, by term, the effect on term GPA from signed grading. The cell
numbers represent the number of students who would see a gain, loss, or unchanged value in
their term GPA due to signed grading.
Overall Change In Student Term GPA From Signed Grades
Term GPA Gain
Spring 2005
1,641
13.2
4,424
35.5
6,393
51.3
12,458
301
10.1
851
28.7
1,816
61.2
2,968
1,753
13.0
4,553
33.7
7,191
53.3
13,497
Winter 2006
46
9.7
80
16.9
346
73.3
472
Spring 2006
1,642
13.7
3,945
33.0
6,369
53.3
11,956
233
8.9
708
27.1
1,675
64.0
2,616
5,616
12.8
14,561
33.1
23,790
54.1
43,967
Summer 2006
All
%
N
%
All
N
Fall 2005
N
Term GPA Unchanged
Term
Summer 2005
%
Term GPA Loss
N
Table 8. Term GPA Gain and Loss Percentages
Data in Table 8 show that during the course of this trial, more students would see
losses in their term GPA than would see gains (better than two to one). The actual statistical
values for change (difference between signed and traditional grading) in term GPA are
displayed in Table 9. On average, signed grading caused term GPAs to decrease
approximately -.025 points.
Page 9 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Mean Change In Term GPA Using Signed Grading
Term
Mean
Std
Median
N
Spring 2005
-.025
0.079
0.000
12,458
Summer 2005
-.039
0.136
0.000
2,968
Fall 2005
-.022
0.075
0.000
13,497
Winter 2006
-.024
0.167
0.000
472
Spring 2006
-.020
0.075
0.000
11,956
Summer 2006
-.037
0.132
0.000
2,616
-.025
0.088
0.000
43,967
Overall
Table 9. Statistical Changes In Term GPAs
Tables 8 and 9 clearly point out that signed grading has the effect of lowering term
GPAs of students, and by extension, cumulative GPAs in the same way.
To examine the effect of signed grading on cumulative GPA, a sub-group of students
who received signed grades was selected. The sub-group consisted of 6,939 students who
were enrolled full-time in the spring 2005, fall 2005 and spring 2006 semesters (three
consecutive terms). Selecting students who were enrolled full-time continuously over the
three terms provided a sub-group with the greatest probability of having the most signed
grades and therefore their having a cumulative GPAs more influenced by signed grading.
Table 10 displays the change in the sub-group cumulative GPA as a result of signed
grading. If adopted, the majority of students (57 %) would see a loss in their cumulative
GPA, 16% would see a gain, and 26% would see no difference.
Change In Cumulative GPA Using Signed Grades
Cumulative GPA
Gain
Sub-Group
Cumulative GPA
Loss
Cumulative GPA
Unchanged
All
N
%
N
%
N
%
N
1,132
16.31
3,992
57.53
1,815
26.16
6,939
Table 10. Cumulative GPA Gain and Loss Percentages
Page 10 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Table 11 displays the statistical changes in cumulative GPAs of the sub-group.
Students who had a gain in their cumulative GPA, on average, recognized a 0.02 positive
change, while students who had a loss would post a -0.03 change.
Direction Of
GPA
Change
Sub-group Mean Change In Cumulative GPA Using Signed Grading
Mean
Std
Median
N
%
Cumulative
GPA Gain
0.02
0.02
0.01
1,132
16.31
Cumulative
GPA Loss
-0.03
0.02
-0.02
3,992
57.53
Cumulative
GPA
Unchanged
0.00
0.00
0.00
1,815
26.16
All
-0.01
0.02
-0.01
6,939
100.00
Table 11. Statistical Changes In Sub-Group Cumulative GPA
Table 11 displays a slight overall negative change in GPA for the sub-group.
Because the number of signed courses was limited during this trial, the small changes
reported in Table 11 may be a function of the number of signed courses a student
experienced.
To establish the relationship between GPA (both term and cumulative) and the
number of signed courses taken, the average change in GPA's was calculated and grouped on
the number of signed courses a student took. These data are reported in Tables 11b and 11c.
Not withstanding the limited number of terms and courses in this trial, Tables 11b
and 11c confirm that the greater the number of signed courses taken, the larger the negative
mean change in GPA. This trend suggests that if more courses were graded using signed
grades the magnitude of GPA changes would also increase. The negative skewing of GPA's
can easily be seen in Table 3 that shows the number of minus grades awarded consistently
outnumbers the plus grades awarded. Obviously, when this occurs the change in GPA is
going to be more negative than positive.
Page 11 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Sub-group Mean Change In Term GPA Relative To Signed Courses Enrolled
Number of Signed Courses Enrolled
None
Term Mean
Spring
2005 0.000
1
N
2
Mean
N
3
Mean
N
4
Mean
N
5
Mean
N
6+
Mean N Mean N
770 -.012 1,656 -.024 2,039 -.033 1,551 -.038
662 -.046 212 -.052 49
Fall
2005 0.000 1,066 -.010 2,026 -.020 2,011 -.030 1,202 -.040
495 -.048 117 -.066 22
Spring
2006 0.000 1,261 -.010 1,983 -.018 1,891 -.024 1,196 -.029
479 -.044 110 -.022 19
All
0.000 3,097 -.011 5,665 -.021 5,941 -.030 3,949 -.036 1636 -.046 439 -.049 90
Table 11.a Mean Change In Sub-Group Term GPA Relative To Signed Courses Enrolled
Sub-group Mean Change In Cumulative GPA Relative To Signed Courses Enrolled
Number of Signed Courses Enrolled
None*
1
Mean
2
N
Mean
3
N
Mean
4
N
Mean
5
N
6+
Mean
N
Mean N Mean N
2005
-.005
770 -.010 1,656 -.012 2,039 -.014 1,551 -.014
662 -.015 212 -.023 49
Fall
2005
-.006 1,066 -.009 2,026 -.013 2,011 -.015 1,202 -.016
495 -.019 117 -.029 22
Spring
2006 -.007 1,261 -.010 1,983 -.013 1,891 -.014 1,196 -.017
479 -.019 110 -.015 19
Term
Spring
All
-.006 3,097 -.010 5,665 -.012 5,941 -.014 3,949 -.015 1636 -.017 439 -.023 90
Table 11.b Mean Change In Sub-Group Cumulative GPA Relative
To Signed Courses Enrolled
* Values for the mean are not zero due to cumulative GPA's calculated under study
limitation #1.
Page 12 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Table 12 presents the net change in cumulative GPA of the sub-group by student’s
major college. Students from the Arts and Letters College recognized the greatest number of
GPAs that would post losses, while students in the Community College would see the
greatest number of gains.
Sub-group Change In Cumulative GPA
Cumulative
Cumulative
Cumulative
GPA Gain
GPA Loss
GPA Unchanged
N
%
N
%
N
%
N
Arts and Letters
306
16.81
1,158
63.63
356
19.56
1,820
Community College
66
19.88
177
53.31
89
26.81
332
Education & Behavioral Science
128
14.87
440
51.10
293
34.03
861
Gordon Ford College of
Business
184
17.61
557
53.30
304
29.09
1,045
Graduate Studies
32
11.81
108
39.85
131
48.34
271
Health & Human Services
153
13.91
632
57.45
315
28.64
1,100
Science & Engineering
222
18.38
734
60.76
252
20.86
1,208
University College
41
13.58
186
61.59
75
24.83
302
1,132
16.31
3,992
57.53
1,815
26.16
6,939
All
Student’s Major College
All
Table 12. Gain/Loss Percentages In Sub-group Cumulative GPA By College
Page 13 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Section IV: Effect Of Signed Grading of on Specific Student Groups
In analyzing the influence of signed grading on the following groups, it should be
noted that the criteria for each group is based on cumulative GPA. In most cases, cumulative
GPA includes semesters of traditional grading that occurred prior to the study trial. As such it
is problematic to estimate the full effect of signed grading on cumulative GPA and in turn,
the influence of signed grading on each of the following sub-groups.
--Students With 4.0 GPAs (Sub-group)
In terms of quality points, the proposed signed grading scheme (Table 1) weights an
“A+” and an “A” the same while an “A-” earns fewer quality points. This has the net effect
of reducing the number of students who can achieve a cumulative GPA of 4.0.
Table 13 illustrates the change in number of students (from the sub-group of 6,939
students) who have a 4.0 GPA with traditional grading, and the number who retain that 4.0
when signed grading is applied to the GPA calculation.
Students Whose Cumulative GPA Is 4.0 Without Signed Grading
Students Whose Cumulative GPA Is 4.0 With Signed Grading
Students Whoso GPA Is 4.0 Without Signed Grading But Less
Than 4.0 With Signed Grading
161
102
(63%)
59
(36%)
Table 13 Effects of Signed Grading On 4.0 Cumulative GPAs
--Candidates for Graduation (Sub-group)
Because the actual number of signed grades a student received in the trial period was
limited, the full effect of signed grading on student graduation cannot be accurately
measured, as most of the courses contributing to graduation GPA are traditional grades.
--Students Receiving Financial Aid (Sub-group)
Table 14 summarizes the effect of signed grading on student financial aid eligibility.
Financial aid eligibility is based on cumulative GPA. Using the previously defined sub-group
the results of signed grading on financial aid eligibility are reported in Table 14. Overall, the
majority of students had no change in their financial aid eligibility status as a result of signed
grades. Only five students would loose eligibility (<1%) as a result of signed grading.
Page 14 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Sub
Group
Number losing
financial aid with
signed grading but
not with
traditional grading
Number losing
financial aid
with traditional
grading but not
with signed
Number
losing aid
regardless
of grading
method
Number who
would keep
aid regardless
of grading
method
Total In
Financial
Aid
Sub-group
5 (<1%)
0
104 (9.1%)
1,031 (90.4%)
1,140
Table 14. Effect of Signed Grading on Financial Aid Eligibility (Cumulative GPA)
-- Student Athletes (Sub-group)
Table 15 summarizes the effect of signed grading on student athletes. Eligibility for
this group was defined as having a cumulative GPA of at least 2.0. (Student athletes on
scholarship must sustain satisfactory progress based on their classification. The 2.0 GPA
value established here is a proxy for satisfactory progress.) Using the previously defined
sub-group, selecting only athletes, is reported in Table 15. Most athletes (92%) had GPAs of
2.0 or better regardless of grading method. A small number (17) had GPAs that fell below
2.0 regardless of grading method. Using signed grading, only 1 student athlete was
identified that would have a GPA below 2.0.
Term
Sub
Group
Number losing
eligibility with
signed grading but
not with
traditional grading
1 (<1%)
Number losing
eligibility with
traditional
grading but not
with signed
Number losing
eligibility
regardless of
grading method
0
17 (7%)
Number who
would keep
eligibility
regardless of
grading
method
215 (92%)
Table 15. Effect of Signed Grading on Student Athletes (Cumulative GPA)
Page 15 of 19
Total In
Athletes
Sub-group
233
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
-- Honors Students (Sub-group)
Honors students must maintain a 3.2 to be eligible to participate in honors. This group
had no students loosing honors eligibility as a result of signed grading.
Term
Sub
Group
Number losing
eligibility with
signed grading but
not with traditional
grading
0
Number losing
eligibility with
traditional
grading but not
with signed
Number losing
eligibility
regardless of
grading method
0
0
Number who
would keep
eligibility
regardless of
grading
method
Total In
Honors
Subgroup
154 (100%)
154
Table 16. Effect of Signed Grading on Honors Eligibility (Cumulative GPA)
Page 16 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Section V: Grade Distribution Comparison
To estimate the overall influence of signed grading on grade inflation, the distribution
of letter grades from the spring, summer and fall terms of 2003 and 2004 were contrasted
with the signed grades from the trial period. These data are summarized in Table 17. The
shaded rows indicate the grading trial terms. (Grades have been stripped of their signs.)
Data from this table suggest that signed grading decreased the number of "As"
awarded in the spring, summer and fall terms, while modestly increasing the number of "Bs"
awarded. The distribution of "C", "D", and "F" grades appears consistent with previous
terms. Based on this trial, signed grading does not appear to increase grade inflation.
Letter Grade Distribution
A
B
C
D
F
%
%
%
%
%
Spring 2003
40.50
29.53
17.32
5.70
6.95
Spring 2004
Spring 2005
40.69
37.63
29.54
30.12
17.36
18.12
5.77
6.28
6.64
7.84
Spring 2006
38.04
31.15
17.81
5.95
7.05
Summer 2003
57.81
25.92
10.68
3.04
2.56
Summer 2004
57.65
25.26
11.35
3.00
2.73
Summer 2005
51.43
28.80
13.46
3.08
3.22
Summer 2006
51.06
29.28
13.29
3.12
3.25
Fall 2003
39.05
29.96
18.06
5.98
6.95
Fall 2004
39.60
29.07
17.34
6.12
7.87
Fall 2005
37.19
30.36
18.36
6.36
7.74
Term
Table 17. Grade Distributions By Term (Shaded cells indicate grading trial period)
Page 17 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
Section VI: Study Summary
•
Over the six terms of this grading trial, faculty participation was consistently below
50% with no more than 32% of the faculty electing to use signed grading exclusively.
The greater part of the faculty continued to use traditional grading throughout the trial
period.
•
With the exception of Winter Term 2006, the distributions of signed grades were
similar through the trial period. In sections that had signed grades, about 60% were
not signed, 17% were plus signed, and about 24% negatively signed. "A-" grades
out numbered "A+" grades by nearly 2 to 1, "B-" grades outnumbered "B+" grades by
about 25% while pluses and minus at the "C" level were about equal. College grade
distributions were similar.
•
Because minus grades consistently outnumbered plus grades, and because an "A+"
grade carried the same quality points as a regular "A" grade, term GPAs, computed
with signed grading, tended to be lower than non-signed grades.
•
In comparing term GPAs calculated with and without signed grades it was found that
signed grading harms more students than it helps. About 33% of all term GPAs
would be lower as a result of signed grading, while about 13% would be higher. The
average difference seen during this trial was about -0.025 grade points. However,
further analysis demonstrated that the more signed courses a student took, the larger
the average negative GPA difference.
•
Measuring the full influence of signed grading on selected student groups was
severely limited by the necessity of using cumulative GPA's. Cumulative GPA
includes semesters of traditional grading that occurred prior to the study trial. The
limited number of signed-graded courses a student may have taken further
compounds this problem. Including non-signed terms and courses greatly reduced
the ability to measure the effect of signed grading on the sub-groups:
o Data from the sub-group (students enrolled full time in the Spring of 2005,
Fall 2005 and Spring 2006) show that majority of students (about 58%) would
have a cumulative GPA loss while only 16% would see a GPA gain. The
average difference in cumulative GPA was small at -0.01. There was
insufficient data to determine the full extent of signed grading on cumulative
GPAs.
o During the course of this trial, about 36% of the students who had a
cumulative 4.0 average with traditional grades would lose it with signed
grades.
o There was insufficient data to determine if candidates for graduation would be
affected by signed grades.
Page 18 of 19
Plus / Minus Grading Trial
Office Of Institutional Research Fall 2006
o Less than 1% of the students in this study would have lost financial aid due to
signed grading.
o Less than 1% of the student athletes in this study would have lost eligibility
due to signed grading.
o Signed grading had no effect on students participating in honors.
•
Signed grading did not appear to influence grade inflation but to some extent had the
opposite effect by reducing the number of "As" awarded.
•
Data from this trial point to the fact that the overall effect of signed grading on
student GPA is negative. However, the relatively low number of sign-graded courses
in this trial severely limits the full extent to which the effects of signed grading can be
gauged.
Page 19 of 19